DESCRIPTION: Orientation selectivity in the mammalian primary visual cortex is a fundamental feature of cortical organization and function and is likely to provide a key to understanding the general rules of neocortical information processing. The original model for orientation selectivity, proposed by Hubel and Wiesel, in which orientation selectivity is based on the convergence of thalamic afferents, has received considerable experimental support. However, subsequent theories and experiments suggest that local cortical circuits, especially inhibitory connections, may also be critical. Little is known, however, about the structural and functional organization of inhibition as it relates to the organization of orientation columns in cortex. Two of the major aims of this proposal are to examine these relationships in ferrets, using a combination of in vivo imaging techniques to define orientation domains, and in vitro brain slices to probe the functional and structural of inhibitory circuits relative to these domains. Experiments using laser photostimulation, voltage-sensitive dye recordings, and intracellular staining are proposed to determine whether inhibition arises primarily form columns of the same or different orientations. Further photostimulation and voltage-sensitive dye imaging experiments are designed to examine whether one component of columnar circuitry, the layer 6 to 4 projection, could selectively amplify the orientation selectivity that is created by thalamic afferents. While patterning of thalamic inputs probably plays a fundamental role in generating orientation selectivity, how such patterning is achieved remains an open question. The role of spontaneous retinal activity in generating orientation selectivity will be tested by using new stimulation techniques to manipulate the levels and patterns of neural activity emanating from the retina during early postnatal development, and assessing the impact of these manipulations on the organization of orientation columns and orientation selectivity in cortex. Finally, as the approach of laser photostimulation has the potential for greatly accelerating the analysis of local circuits in the cortex and elsewhere, a series of experiments are proposed to develop a new, more general method of photostimulation, based on photolabile liposomes. This will enable researchers to deliver bioactive molecules with unprecedented spatial and temporal resolution. Understanding the basic computational circuits in cortex is fundamental to understanding the basis of numerous pathological states of cortical organization, including amblyopia and mental retardation, while determining the role of early spontaneous activity in constructing circuits has important implications for the effects of toxins and drugs during prenatal development in humans.